基于LSTM神经网络修改参考输入的数控系统轨迹跟踪控制  被引量:1

LSTM Neural-network Based Reference Modification for Trajectory Tracking of CNC Systems

在线阅读下载全文

作  者:王小东 王柱 林承敏 王天雷 黄桂武 周智恒[3] Wang Xiaodong;Wang Zhu;Lin Chengmin;Wang Tianlei;Huang Guiwu;Zhou Zhiheng(Guangdong Kejie Mechanical Automation Co.,Ltd.,Jiangmen,Guangdong 529030,China;Faculty of Intelligent Manufacturing,Wuyi University,Jiangmen,Guangdong 529020,China;School of Eectronics and Information Engineering,South China University of Technology,Guangzhou 510640,China)

机构地区:[1]广东科杰机械自动化有限公司,广东江门529030 [2]五邑大学智能制造学部,广东江门529020 [3]华南理工大学电信工程学院,广州510640

出  处:《机电工程技术》2022年第12期52-56,155,共6页Mechanical & Electrical Engineering Technology

基  金:2020年广东省科技专项资金(“大专项+任务清单”)项目;2021年度江门市重大科技计划项目(编号:江科[2021]181号)。

摘  要:为减少数控加工系统的加工误差,提出了一种基于长短时记忆神经网络(LSTM-NN)的计算机数控机床轨迹跟踪控制参考轨迹修正方法。该方法的实现是通过使用LSTM-NN建立实际轨迹与伺服系统参考轨迹之间的映射,并在不改变控制器的情况下修改参考轨迹以补偿跟踪误差。具体而言,首先分析数控系统伺服系统的结构,并建立单轴数学模型。其次,根据系统模型选择LSTM-NN的输入特征,运用伺服系统的实际轨迹和参考轨迹来训练LSTM-NN。最后生成可根据参考轨迹修改输入的模型。相较于三环PID控制伺服系统与LSTM神经网络逆模型控制系统的轮廓控制,所提出的方法在数控系统运行后的轮廓误差降低了11μm,有效提高了加工精度。In order to reduce the machining error of CNC machining system,a reference trajectory correction method based on long and short term memory neural network(LSTM-NN)was proposed.The method was implemented by establishing the mapping between the actual trajectory and the servo system reference trajectory using LSTM-NN,and modifying the reference trajectory to compensate the tracking error without changing the controller.Specifically,firstly,the structure of servo system was analyzed and a simplified uniaxial model was established.Secondly,the input features of the LSTM-NN was selected according to the system model,and trained the LSTM-NN using the actual and reference trajectories of the servo system.Finally,a model was generated that could modify the input according to the reference trajectory.Compared with the contour control of the three-loop PID control servo system and LSTM neural network inverse model control system,the contour error of the proposed method is reduced by 11μm after the operation of the CNC system,and the machining accuracy is effectively improved.

关 键 词:计算机数控(CNC) LSTM-NN 跟踪控制 

分 类 号:TP18[自动化与计算机技术—控制理论与控制工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象